NEUROLOGY AND PRECLINICAL NEUROLOGICAL STUDIES - ORIGINAL ARTICLE

Emerging issues on selection criteria of levodopa carbidopainfusion therapy: considerations on outcome of 28 consecutivepatients

Mariachiara Sensi • F. Preda • L. Trevisani • E. Contini •

D. Gragnaniello • J. G. Capone • E. Sette • N. Golfre-Andreasi •

V. Tugnoli • M. R. Tola • R. Quatrale

Received: 27 August 2013 / Accepted: 25 December 2013

� Springer-Verlag Wien 2014

Abstract Many studies confirmed the efficacy and safety

of continuous infusion of intrajejunal levodopa/carbidopa

gel (CIILG) for advanced Parkinson’s disease (PD).

Although this treatment is widely used, definite inclusion/

exclusion criteria do not exist. In this prospective open-

label study, we evaluated the long-term outcome in 28

consecutive patients and sought to detect any predictive

factor to identify the best candidates for CIILG therapy.

The assessment was carried out routinely at baseline, after

6 months and every year with UPDRS III–IV, FOG

Questionnaire, non-motor symptoms scale, PD question-

naire (PDQ-8), cognitive and psychiatric status evaluation

(MMSE, FAB, NPI) and caregiver’s quality of life. 17/28

patients reached the 24-month follow-up. A statistically

significant beneficial effect was shown on motor compli-

cations in short- and long-term follow-up, also on axial

symptoms like gait disturbances. A concomitant improve-

ment in PDQ8 score was observed, with a parallel mild

amelioration, but not significant, on Caregivers QoL. When

classified according to their outcome on QoL, the only

predictive positive factor was less severe at Neuropsychi-

atric Inventory (NPI) score at baseline. Considering the

improvement in motor scores (duration of ‘‘off’’ period),

the more advanced age was associated with a poorer out-

come. Our results confirmed a sustained efficacy and safety

in long-term follow-up and suggest that younger age at

operation and absence or mild presence of psychiatric/

behavioural symptoms could be considered valid predict-

ing factors in selecting the best candidates for this effica-

cious therapy.

Keywords Parkinson disease � Duodopa � Selection

criteria

Introduction

Continuous infusion of Levodopa Carbidopa Intestinal

Gel (LCIG, Duodopa�) is broadly recognized as a ben-

eficial treatment for advanced and/or complicated PD.

Since the first experiences on small randomized studies

(Kurth et al. 1993; Nyholm et al. 2003), until the more

recent prospective ongoing studies (Palhagen et al. 2012;

Fernandez et al. 2013), its efficacy on motor and non-

motor fluctuations and also on quality of life is fully

demonstrated. More than 25 works are now present in

literature exploring many dimensions of PD outcome, but

the variability of patients selected for this therapeutic

option is evident, as underlined by the recent review from

(Nyholm 2012).

Unlike the clear-cut selection criteria for Deep Brain

Stimulation (DBS) that are well defined since the first

experiences (Defer et al. 1999), after almost 10 years of

practice accurate indications for LCIG infusion are still

M. Sensi (&) � F. Preda � E. Contini � D. Gragnaniello �J. G. Capone � E. Sette � N. Golfre-Andreasi � V. Tugnoli �M. R. Tola

Department of Neuroscience-Rehabilitation, S. Anna University-

Hospital of Ferrara, via A.Moro, 44124 Ferrara, Italy

e-mail: mchiasen@gmail.com

L. Trevisani

Department of Gastroenterology- Digestive Endoscopy, S. Anna

University Hospital of Ferrara, via A.Moro 8, 44124 Ferrara,

Italy

R. Quatrale

Department of Neurology, Ospedale dell’Angelo, via

Paccagnella 11, 30174 Mestre-Venezia, Italy

123

J Neural Transm

DOI 10.1007/s00702-013-1153-3

lacking (Johansson and Nyholm 2012; Antonini and Tolosa

2009; Volkmann et al. 2013).

The urgency to identify inclusion and exclusion criteria

becomes more evident when the cost of this treatment is

taken into account (Valldeoriola et al. 2013).

The clinical profile of the best candidate to Duodopa is

not well defined yet, and this is crucial considering that this

treatment is often considered as an alternative to DBS. The

population selected is also heterogeneous if we consider

age, concomitant diseases, and cognitive or psychiatric

status. The consequence is that the inclusion criteria are

paradoxically contradictory (Antonini et al. 2008; Devos

2009).

This is quite evident for cognitive status: the only study

that considered dementia as an inclusion criteria is the

French multicenter study (Devos 2009), whereas cognitive

impairment was considered a main exclusion criteria in

many other studies (Antonini et al. 2007; Puente et al.

2010; Fasano et al. 2012; Zibetti et al. 2013). Recently, the

Danish Movement Disorder Society (DANMODIS) and the

Swedish Movement Disorder Society (SWEMODIS)

issued joint Scandinavian Consensus Guidelines for the use

of LCIG infusion (DANMODIS 2008) and proposed that

slight to moderate dementia is not a contraindication,

whereas severe dementia should be considered a relative

contraindication.

Bearing in mind this background, we examined our

population of Duodopa patients, trying to identify which

features contributed to the best outcome and to characterize

the best candidate. Our aim was also to detect any sensitive

tool (among tests/assessments/scales used for motor, non-

motor, axial, cognitive, behavioral, gastroenterological

symptoms and caregiver burden) that might be helpful to

clinicians as predictive factors for good or poor outcome.

Materials and methods

Patients

In this prospective open-label study, we evaluated 28

consecutive Duodopa patients with clinical diagnosis of

PD, 16 men and 12 women, who underwent LCGI infusion

in Ferrara from January 2008 to March 2013. The study

was approved by our ethical committee. Patients were

treated in a clinical routine setting and signed an informed

consent about the study and the use of medical data. All

patients were on oral levodopa treatment before the

intervention.

The inclusion criteria chosen fulfilled the Summary of

Product Characteristics criteria (Idiopathic PD with severe

motor fluctuations and hyper-/dyskinesia when available

combinations of anti-parkinsonian drugs have not given

satisfactory results). The exclusion criteria, chosen mostly

on our previous experiences, were: all form of atypical and

secondary parkinsonism, severe PD-related psychosis and/

or severe hallucinations, other concomitant general dis-

eases that may interfere with the correct compliance with

the device. The unavailability of a caregiver was consid-

ered exclusion criteria in patients with dementia or with

mild cognitive impairment.

Procedure

All patients were switched from their parkinsonian phar-

macological treatment (L-dopa, dopamine agonists, COMT

inhibitors, amantadine and subcutaneous apomorphine) to

continuous intrajejenual L-Dopa/carbidopa gel infusion in

the first day of hospitalization. From days 1 to 3, a tem-

porary nasoduodenal tube was positioned by a PD nurse

specialist in the ward of our neurological department. The

PD nurse trained patients and caregivers (relatives or any

person assisting the patient at home) in handling the

device, whereas nurses of our ward trained them in the

management of the stoma (i.e., medication). Once at home,

all caregivers were able to manage the procedure

autonomously.

On day 4, percutaneous endoscopic gastrostomy with

a jejunal tube (PEG-J) was performed by the gastroen-

terologist under local anesthesia and sedation with a

short acting benzodiazepine. In the following 3–7 days

of hospitalization, LCIG infusion via PEG-J was started

and the dose titrated according to patient’s previous

daily oral intake of L-dopa plus the L-dopa equivalent

dose of dopamine agonists. Long acting dopamine

agonists and/or oral slow-release L-dopa preparations

were used to manage nocturnal motor symptoms in

15/28 patients. The continuous dose of infusion was

titrated individually to reach an optimal clinical

response in steps of 2–4 mg L-dopa/h (0.1–0.2 ml/h).

Monthly follow-up visits were arranged with the neu-

rologist and gastroenterologist in the first 3 months, and

thereafter every 6 months. The tube replacement is

usually scheduled on demand, depending on the wear

conditions, but we observed that it usually takes place

every 9–12 months (range 9–18 months).

Evaluations

All scales/assessments were completed at baseline before

the initiation of intrajejunal LCIG infusion, after 6 months

and every year. All evaluations were performed in ‘‘best

on’’ state. Motor status was evaluated with UPDRS III

(motor) and IV (complications), H&Y stage, FOG ques-

tionnaire. Non-motor symptoms were tested with the

NMSS and health-related quality of life with the PDQ-8

M. Sensi et al.

123

scale. All patients underwent cognitive (MMSE, FAB) and

psychiatric (NPI) status evaluation. The caregiver’s quality

of life was tested using a specific tool: SQLC (Glozman

et al. 1998), which was designed specifically for caregivers

of patients with PD to assess the occupational, social and

leisure activities of the caregiver and their responsibilities

towards patient in everyday life.

We also sought to identify factors associated with or

predicting the best long-term outcome. The following data

were collected and used for analysis: gender, type of PD,

age of onset, PD duration, presence and features of cog-

nitive impairment pre- and post-LICG, doses of Duodopa

profiles.

Statistical methods and analysis

Analysis of variances for repeated measures applying the

Bonferroni correction, with a significance at p \ 0.01 as

adjustment for multiple comparisons, was used to compare

performances assessed by UPDRS and its subitems, H&Y,

FOG, NMS, MMSE, FAB, NPI, PDQ8 and SQLC at

baseline and 6 and 24 months after treatment.

The percentages of improvement were calculated

according to the following formula: (baseline score - fol-

low-up score)/baseline score 9 100.

To investigate whether baseline characteristics could be

considered as predictable factors for quality of life after

24 months of treatment, the whole population was first

divided into two groups according to the percentage of

PDQ8 improvement obtained after 2 years of treatment

[Group 1 (good outcomes) = improvement C30 %. Group

2 (poor outcomes) = improvement \30 %]. One-way

analysis of variance (ANOVA), with PDQ8 improvement

as the independent variable, was performed. Subsequently,

to investigate whether baseline characteristics could be

considered as predictable factors for motor aspects

improvement after 24 months of treatment, the whole

population was divided into two subgroups on the basis of

improvement reached in item 39, after 2 years of treatment

[Group 1 (good outcomes) = improvement C50 %, Group

2 (poor outcomes) = improvement \50 %] and the same

was performed using item 39 improvement as independent

factor.

Statistical analyses were conducted using the SPSS ver.

9.0 statistical package with significance set at p B 0.05.

Safety

Vital signs, extended laboratory data with nutritional pro-

file, physical status, electrocardiogram, and the presence of

polyneuropathy (clinically and by EMG readings) were

evaluated at baseline and every 6 months. Adverse events

were recorded.

Results

Of the 28 PD patients who were enrolled between 2008 and

2013, seven cases had LCIG infusion as second-line

treatment since apomorphine pump (4/28) or STN stimu-

lation therapy had failed (3/28). In ten patients, these two

options were contraindicated. Duodopa therapy was instead

the first-line treatment in 11/28 patients who were also

eligible for DBS or apomorphine pump.

14/28 patients manifested at least one adverse event as

reported in Table 1. Peripheral neuropathy (PN) was dis-

closed in four patients at baseline (mild clinical signs of

PN, associated with electrophysiological mixed motor and

sensory alterations) and at last follow-up was present in

9/28 patients; all the forms were very mild in eight patients

and remained stable after vit. B supplementation. A severe

form of PN was detected in one patient without any clinical

or electrophysiological alteration at baseline. After

5 months of Duodopa treatment, he developed a severe

subacute sensory-motor demyelinating polyneuropathy,

associated with significant alterations in MMA, Hcy and

Vitamin B12 levels. This patient died for Duodopa prob-

ably related problems (malnutrition associated with severe

polyneuropathy) after 6.2 months for nosocomial

pneumonia.

Table 1 Adverse events in 28 consecutive PD-LCIG patients

Adverse events No.pt Presence

at

baseline

Leading to

discontinuation

Related to levodopa

Polyneuropathy 9 4 1

Weight loss 3 0 0

Mood disturbance 2 3 0

Hallucinations 3 3 0

Agitation 3 1 1

Total 20 11 2

Related to procedure (gastrostomy-PEG)

Duodenal ulceration 1 NA 1

Peritonitis 2 NA 2

Peristomal infections 1 NA 0

Total 4 NA 3

Related to device

PEG pulled out accidently 1 NA 0

Dislocation/replacement of

jejunal tube

4 NA 0

Tube occlusion 2 NA 1

Granulation at PEG puncture 4 NA 0

Pump failure 5 NA 0

Total 16 NA 1

Total 40 11 6

NA not applicable

Emerging issues in Duodopa patients’ selection criteria

123

Three patients died for causes unrelated to Duodopa

treatment (pneumonia, myocardial infarction and sepsis)

after a period of 8.8, 30.2 and 2.4 months of infusion,

respectively. In two patients in very advanced stage of

disease (H&Y = 5) with other comorbidity, it was not

possible to perform PEG-J for technical reasons (no

transillumination); consequentially, they were submitted to

jejunostomy but they died in the immediate post-surgery

due to peritonitis. Both of them were submitted to surgery

with peri-operatory prophylaxis with antibiotics (as long as

treatment with proton pump inhibitors), following the

protocol of our Center.

During the study period, 3 patients (10.7 %) discontin-

ued Duodopa infusion after 26, 44 months and 4 years of

follow-up; one for severe psychosis and two for repeated

device problems associated in one case with severe psy-

chosis (Table 1).

Moderate cognitive deterioration or dementia (DSM IV

criteria) was present in 12 patients (MMSE \ 24).

17 patients reached the scheduled follow-up of

24 months evaluation after a mean time period of

32.4 ± 9.4 months (range 24.2–48.1); we report here the

extensive analysis of treatment outcome. The main clinical

characteristics at baseline and at follow-up are summarized

in Table 2.

The mean age was 67.6 years ± 6.1, and the mean BMI

was 23.6 ± 2.9 kg/m2. The mean duration of PD disease was

15.47 ± 4.04 years, and the mean Hoehn and Yahr Score in

the ‘‘On’’ phase was 3.17 ± 0.8. The previous mean total

levodopa equivalent dose (LEDD) was 1158.9 ± 334.5 mg.

The mean levodopa dose at start of infusion was

75.2 ± 27.0 ml/day and remained stable during the follow-

up. Concomitant medications other than Duodopa therapy

were modified during the follow-up, as seen in Table 3. The

percentage of patients receiving levodopa infusion as

monotherapy at last follow-up increased by 23 % compared

to baseline since oral dopamine agonists, subcutaneous

apomorphine, amantadine, COMT inhibitors, and/or oral

nighttime levodopa/carbidopa tablets were reduced. The

use of antipsychotics remained substantially unchanged, but

in three patients they were increased in dosage. There was a

significant increment in vit. B12 e folate considering the

concomitant increase in homocysteine levels and the onset

of mild polyneuropathy in five patients (Table 3).

Data analysis showed that UPDRS III and H&Y in

‘‘On’’ condition did not change after 6 and 24 months

(Table 4); sub evaluation of axial items (18, 28–30)

revealed a significant improvement in gait (item 29) at

6 months (p = 0.012) that returned to basal values at

24 months (Fig. 1).

Total UPDRS IV (items 32–42) showed a significant

improvement at 6 months (p \ 0.0001) that was maintained

at last follow-up (p \ 0.0001). Items 32 (duration of dys-

kinesia) and 33 (degree of disability induced by dyskinesia)

were, respectively, 2.2 ± 1.1, 1.8 ± 1.0, at baseline and

showed an improvement already after 6 months that became

significant after 24 months (p = 0.006 and p = 0.004,

respectively). Item 39 (duration of ‘‘Off’’ periods) improved

consistently at 6 (p = 0.001) and 24 months (p [ 0.0001).

The mean improvement was of 48 % (Fig. 1).

Similar to the improvement in item 29 (gait), the total

score of FOG questionnaire at 6 months (p = 0.001) and

2 years (p = 0.03) is significantly lower than the pre-

treatment condition; however, there is a significant deteri-

oration from 6 months to 2 years, where the score is

Table 2 Main clinical characteristics of 17 patients with long-term

follow-up

Demographics and medical history

Gender M/F (%) 58.8/41.2

Mean age at CIGIL onset (years) 67.6 ± 6.1

BMI (kg/m2) 23.6 ± 2.9

Rigid/akinetic phenotype at onset (%) 41.17

Mean duration of PD at onset (years) 15.47 ± 4.04

Follow-up duration (months) 32.47 ± 9.47

MMSE at onset (\24) % 47.0

FAB at onset (\13) % 52.9

Hallucinations at onset (%) 35.7

Agitation at onset (%) 11.8

Mood disturbances at onset (%) 21.2

Total LEDD pre-CIGIL (mg) 1158.9 ± 334.5

FAB Frontal assessment battery, MMSE mini-mental state examina-

tion, LEDD levodopa equivalent daily dose

Table 3 Concomitant medications, used at least by 5 % of patients,

at starting of LCIG and at 6 and 24 months of follow-up (reported as

% of patients)

Baseline 6 months Last

follow-up

Mean LCIG dose 75.2 ± 27 77.8 ± 21 74.8 ± 25

Duodopa monotherapy (%) 41.17 52.94 64.70

Vitamin B12 and folic

acid (%)

0 11.76 52.94

Anxiolytics (%) 29.41 47.91 42.65

Antidepressant (%) 48.5 48.5 36.5

Antipsychotic (%) 17.64 17.64 17.64

Anti-dementia drugs (%) 5.00 5.00 5.00

COMT inhibitors (%) 47.05 23.51 5.00

Amantadine (%) 35.29 11.76 0

Dopaminoagonists (%) 76.47 52.94 35.29

Levodopa–carbidopa tablets at

night time (%)

70.50 50.02 41.17

LCIG levodopa/carbidopa intestinal gel

M. Sensi et al.

123

significantly higher (p = 0.04). Significant improvement

was documented for each subitem (1–6) of the question-

naire (Fig. 2).

In NMS, we have not found significant difference

between the score at follow-up and the score in the baseline

condition (Table 4). The same result is confirmed for all

subitems.

Regarding cognitive status, MMSE showed a progres-

sive and significant, although mild, worsening at 2-year

follow-up (p = 0.011), whereas the mean FAB score

remained stable at 6 and 24 months of follow-up (Table 4).

In PD-demented patients, the mean decline in MMSE at

2 years was of 2.2 points (from 22.4 ± 0.9 at baseline to

20.2 ± 2.6 at last follow-up) while in non-demented PD

patients MMSE remained stable at 2 years of follow-up

(from 27.0 ± 1.1 to 26.5 ± 2.2).

Referring to behavioral symptoms assessed through

NPI, we observed a low total score at baseline (mean 15.7;

range 4–40) that remained low at last follow-up (mean

16.1; range 0–44) (Table 4).

Table 4 Baseline characteristics and follow-up of motor- and non-motor scores (mean ± SD) in 17 patients

Baseline FU 6 months FU 24 months p value (baseline-

FU 6 months)

p value (Baseline-

FU 24 months)

UPDRS III ON 35.5 ± 11.5 33.4 ± 10.8 34.7 ± 12.4 NS NS

H&Y 3.2 ± 0.7 3.1 ± 0.8 3.0 ± 0.8 NS NS

Item 18 1.7 ± 1.2 1.7 ± 1.3 1.6 ± 1.2 NS NS

Item 28 2.1 ± 0.9 1.9 ± 0.9 1.9 ± 1.0 NS NS

Item 29 2.4 ± 0.8 2.0 ± 0.9 2.0 ± 1.0 p = 0.012 NS

Item 30 1.7 ± 0.9 1.6 ± 1.2 1.9 ± 1.3 NS NS

Item 32 2.2 ± 1.0 1.8 ± 1.0 1.2 ± 1.0 NS p = 0.006

Item 33 1.8 ± 1.0 1.4 ± 1.0 1.0 ± 0.9 NS p = 0.004

Item 39 2.3 ± 0.9 1.2 ± 0.6 1.0 ± 0.6 p = 0.001 p \ 0.00001

UPDRS IV 8.4 ± 2.5 5.6 ± 2.7 4.4 ± 1.9 p \ 0.00001 p \ 0.00001

NMS 51.8 ± 37.3 44.6 ± 25.6 38.0 ± 24.7 NS NS

FOGQ 9.9 ± 3.8 4.8 ± 3.7 7.4 ± 5.1 p = 0.001 p = 0.03

MMSE 25.0 ± 2.7 24.4 ± 2.8 23.2 ± 4.1 NS p = 0.011

FAB 13.4 ± 2.9 13.0 ± 2.9 12.3 ± 3.1 NS NS

NPI 15.7 ± 10.1 14.9 ± 8.6 16.1 ± 12.2 NS NS

PDQ 8 46.3 ± 13.7 32.5 ± 16.4 29.9 ± 17.0 p = 0.011 p = 0.006

SQLC 87.8 ± 19.5 91.7 ± 18.9 94.4 ± 20.3 NS NS

FAB Frontal assessment battery, FOG freezing of gait questionnaire, MMSE mini-mental state examination, NMSS non-motor symptoms scale,

NPI neuropsychiatry inventory, PDQ-8 Parkinson’s disease questionnaire, SQLC scale of quality of lifer of caregivers, UPDRS unified Par-

kinson’s disease rating scale, NS not statistically significant

* p \ 0.05

item

29

UPDRS IV

item

32

item

33

Item

39

0

2

4

6

8

10

12

Mea

n U

PD

RS

sco

re

**

* ***

*

B 6 24 B 6 24 B 6 24B 6 24 B 6 24

Fig. 1 Significant outcome in

motor scores of UPDRS (III and

IV) at Baseline (B), after

6 months (6) and 24 months

(24) of follow-up in 17 LCIIG

patients. Item 29 gait, item 32

duration of dyskinesias, item 33

dyskinesias disability, item 39

duration of ‘‘off’’. *p \ 0.05

Emerging issues in Duodopa patients’ selection criteria

123

The mean PDQ-8 score was 46.32 ± 13.6 at baseline

and was significantly reduced at 6 (32.52 ± 16.4)

(p = 0.011) and 24 months (29.9 ± 17.0; p = 0.006). The

mean improvement was 34 % at 6 months and 33 % at last

follow-up (Table 4).

The scale of QoL of Caregiver (SQLC) showed a severe

burden at baseline with a mild but not significant

improvement at 24 months (Table 4). The same result is

repeated for all subitems.

In the same population of 17 patients with long-term

follow-up, we examined the possible presence of predic-

tive factors. In this analysis, we arbitrary chose a cut-off

of more than 30 % of improvement in PDQ 8 and iden-

tified two subgroups of seven (good outcome) and ten

patients (poor outcome). The analysis of the baseline

features of these two samples did not reveal significant

differences except for NPI: the score of the good outcome

group is significantly lower (median score of 8.00 ± 2.8)

than the one of the poor outcomes group (median score of

19.2 ± 11.1). In the former group, a prevalence of

symptoms like anxiety and depression is noted; in the

latter group, agitation and disinhibition prevailed

(Table 5).

Item 39, FOG and H&Y scores were all lower in the

good outcome group compared to those in the group of

poor outcomes, but the difference is not statistically sig-

nificant (Table 5). This may be due to the fact that these

scales have very strict scores and the sample size is not

sufficient to show a meaningful statistical analysis of these

trends.

For the motor status, we analyzed also the improvement

of item 39 (duration of ‘‘Off’’) as a predictive factor

([50 % of improvement) and two distinct populations

emerged: 12 patients with good outcome and five with poor

outcome. In the group of good responders, the age

(\70 years) was significantly lower compared to the group

of poor outcomes (p = 0.039) (Table 5).

Discussion

Our population shares similar results in several outcome

measurements as reported in previous studies.

According to preceding works, ‘‘On’’ scores of UPDRS

parts III were not significantly reduced after 6–24 months

(Palhagen et al. 2012; Antonini et al. 2008; Puente et al.

2010).

The duration and severity of dyskinesias and the dura-

tion of ‘‘Off’’ periods assessed by the UPDRS IV items 32,

33 and 39 were all significantly reduced. The reported

benefits are consistent with the results of previous shorter

clinical trials and case series (Devos 2009; Puente et al.

2010; Nyholm et al. 2005; Eggert et al. 2008; Honig et al.

2009).

Among the axial symptoms, only gait disturbances (item

29 and FOG questionnaire) showed a significant

improvement (including freezing, start and turn hesitation).

These data have been reported hitherto only in Devos and

Nyholm case studies (Devos 2009; Nyholm et al. 2008),

even if it tends to become less evident in long-term follow-

up. As previously described (Zibetti et al. 2013), this

evolution might correlate with the concomitant degree of

cognitive decline observed in long-term follow-up as stated

by the concomitant worsening of MMSE.

Compared to previous studies (Devos 2009; Puente et al.

2010; Fasano et al. 2012; Honig et al. 2009, Reddy et al.

2012), there was a trend in amelioration in all sub items of

NMS but none was significant. Interestingly, the score of

each subitem (except for items of mood, perceptual prob-

lems and attention) was globally mild. Recently,

FOG-Q to

tal 1 2 3 4 5 6

0

1

2

3

4

5

6

7

8

9

10

Mea

n sc

ore

* ** *

B 6 24 B 6 24B 6 24 B 6 24B 6 24 B 6 24B 6 24

*

*

* ** * *

Fig. 2 Outcome of FOG-Q

total and in each subitem at

baseline (B), after 6 months (6)

and 24 months (24). 1 walk, 2

ADL, 3 freezing, 4 freezing

duration, 5 start hesitation, 6

start hesitation duration.

*p \ 0.05

M. Sensi et al.

123

(Chaudhuri et al. 2013) reported that patients with milder

motor disorder have considerable non-motor burden com-

pared to the ones in more advanced stages of disease. This

could probably explain why in our patients, having more

severe motor symptoms, the entity of NMS disturbances

was less troublesome and with little margin of

improvement.

As far as cognitive status is concerned, our population is

characterized by a considerable number of patients mostly

selected in the first years, with a mean lower score at

baseline (8/17 patients with MMSE less than 24; and 9/17

with FAB score lower than 13) as described in first studies

of Devos (2009), Nyholm et al. (2008) and more recently

(Zibetti et al. 2013). Nevertheless, our patients showed a

slower progression of cognitive deterioration both in

demented and non-demented PD, if compared to a similar

population (Zibetti et al. 2013) or referring to PD patients

on oral medications, as reported by Aarsland et al. (2004),

who documented a higher annual mean decline in MMSE

in both populations. Therefore, even if it is untimely to

Table 5 Predictive factors for

QOL and motor improvement in

17 patients with long-term

follow-up

D PDQ8 2 years FU C 30 %

improvement of PDQ8 [ 30 %

after 2 years of follow-up, DItem 39 2 years FU C 50 %

improvement of item 39 (off

duration) [ 50 % after 2 years

of follow-up

* p \ 0.05

Bold values indicate significant

predictive factors

N number of patients

Group 1 = (good outcomes)

D PDQ8 2 years FU C30 %

Group 2 = (poor outcomes)

DPDQ8 2 years FU \30 %

p value

N 7 10

Age 68.3 ± 5.6 67.2 ± 6.8 0.445

Follow-up duration (months) 33.2 ± 9.2 29.03 ± 8.03 0.345

Disease duration (years) 15.9 ± 5.3 15.2 ± 3.2 0.543

Males 4 6 –

Rigid/akinetic phenotype at onset 3 4 –

Total LEDD pre 1146.0 ± 420.8 1168.0 ± 284.0 0.449

Levodopa LEDD pre 852.4 ± 387.3 835.0 ± 231.0 0.454

Dopamino agonist LEDD pre 293.6 ± 137.2 333.0 ± 204.4 0.331

UPDRS III ON baseline 33.0 ± 11.3 37.3 ± 12.0 0.425

UPDRS IV baseline 8.7 ± 2.7 8.1 ± 2.7 0.637

H&Y ON baseline 2.8 ± 0.4 3.4 ± 0.9 0.096

FOG ON baseline 3.4 ± 1.1 4.9 ± 4.8 0.444

MMSE baseline 24.9 ± 2.3 25.0 ± 3.4 0.745

FAB baseline 13.1 ± 3.9 13.6 ± 2.2 0.404

NPI baseline 8.0 – 2.8 19.2 – 11.1 0.021

Item 39 ON baseline 1.7 ± 0.5 2.4 ± 1.0 0.106

Group 1 = (good outcomes)

D Item 39 2 years

FU C 50 %

Group 2 = (poor outcomes)

D Item 39 2 years

FU \ 50 %

p value

N 12 5

Age 67.1 – 7.2 69.0 – 2.5 0.039

Follow-up duration (months) 32.7 ± 9.5 32.0 ± 10.6 0.453

Disease duration (years) 15.7 ± 4.8 15.0 ± 1.4 0.753

Males 7 3 –

Rigid/akinetic phenotype at onset 6 3 –

Total LEDD pre 1115.7 ± 349.9 1262.8 ± 303.0 0.569

Levodopa LEDD pre 812.8 ± 322.1 912.8 ± 226.3 0.432

Dopaminoagonist LEDD pre 302.9 ± 167.5 350.0 ± 211.7 0.567

UPDRS III ON baseline 33.8 ± 12.6 39.6 ± 7.9 0.467

UPDRS IV baseline 8.8 ± 2.8 7.8 ± 1.9 0.577

H&Y baseline 3.2 ± 0.8 3.1 ± 0.5 0.420

FOG baseline 10.2 ± 3.2 9.4 ± 5.1 0.474

MMSE baseline 25.6 ± 2.8 23.4 ± 2.0 0.895

FAB baseline 13.2 ± 3.0 13.7 ± 2.8 0.756

NPI baseline 15.2 ± 10.3 17.3 ± 10.4 0.064

PDQ8 45.3 ± 11.8 48.8 ± 18.8 0.226

Emerging issues in Duodopa patients’ selection criteria

123

conclude that continuing levodopa infusion improves

cognitive functions, as described in two patients by San-

chez-Castaneda et al. (2010), it could be argued that this

therapy does not have detrimental effect on them.

The questionnaire for caregiver burden in PD showed

that the severe caregiver’s stress at baseline did not

improve significantly in short- and long-term follow-up;

the analysis of subitems showed a mild improvement in

social and leisure activities even though the responsibilities

of the caregiver helping the patient in everyday living

remained unchanged. The major burden was secondary not

to the management of the pump but to the constant

supervision needed by these patients.

These results are in contrast with Santos-Garcia et al.

(2012), who reported a great improvement in caregiver

stress and burden and are more in line with the results of

Fasano et al. (2012). These contradictory results might

depend on the difference in sample size, duration of fol-

low-up and in the scale itself.

Peculiarly, when predicting factors were analyzed, sex,

stage, PD duration, total LEDD, Dopamine agonists dos-

age, presence and features of cognitive impairment or

motor phenotype were not predictive of a good or bad

outcome. Few studies tried, in retrospective manner, to

discover some predictive factors: Westin et al. (2006)

reported that the more severe the UPDRS III at baseline,

the better is the outcome, whereas Fasano et al. (2012)

identified a low severity of UPDRS III and GFQ at baseline

as a good predictive factor.

Not surprisingly, considering the significant number of

patients with mild to moderate cognitive decline at base-

line, the better outcome in QoL was reported in the group

with the lower NPI score at baseline. This might suggest

that, regardless of cognitive and motor impairment at

baseline, the presence of clusters of positive symptoms like

disinhibition, hallucinations and agitation that were the

preeminent features of poor outcomes, is a negative

prognostic factor in long-term follow-up. On the other

hand, it is well known that the absence of negative

symptoms, such as apathy or depression, in good outcomes

correlates with a better resilience and a better QoL (Ro-

bottom et al. 2012).

This finding matched with the better outcome in QoL of

patients with milder severity of disease (lower H&Y stage

and off duration, less axial symptoms).

Another important issue that emerged in good outcome

group is the age: the younger the patients, the more sig-

nificant was the improvement in the short- and long-term

impact on motor outcome measured by item 39.

Despite the open design, the small case series and the

lack of a control group of our study, we think that globally

our results provide some suggestive considerations. Firstly,

if behavioral problems seem to be more relevant than

cognitive decline solely as predictive factor (Nyholm et al.

2008), NPI could be considered a valid tool in discrimi-

nating poor versus good outcome. Despite the high prev-

alence of psychiatric symptoms in patients with cognitive

deterioration (Aarsland et al. 2007) and in non-demented

PD patients (Kulisevsky et al. 2008), this assessment is

rarely used routinely in screening phase (Fasano et al.

2012).

Secondly, after evaluating all adverse events, the rate of

discontinuation is lower compared to previous studies

(Zibetti et al. 2013; Nyholm et al. 2008; Nyholm 2012). In

our three patients, the presence of behavioral problems

related to progression of PD was an important contributing

reason for discontinuation, as reported in Devos (2009) and

Antonini et al. (2013) case studies. Hence it becomes

decisive to prevent this complication.

As outlined recently (Hindle 2013), in patients with very

advanced PD, the first attempt to control the behavioral

disturbances is to reach an oversimplification of the ther-

apy, with a significant reduction of concomitant anti-PD

medications. In second instance, where hallucinations are

the preeminent symptoms, atypical anti-psychotic agents

are useful (Nyholm 2012). This was also our experience

throughout the years and this approach permitted a better

compliance to this treatment from patients and caregivers.

Moreover, six patients died (21 %) during the LGCI

infusion but all of them were in very advanced stage

(H&Y [ 4) and old ([70 years) and two already bedrid-

den, so the severity of the disease and the probable

comorbidity were causal factors.

These observations imply some crucial evaluations in

selection criteria, taking into account also that the occur-

rence of behavioral symptoms has an obvious consequence

on caregiver stress and could be a cause of drop-out in the

long-term follow-up (Aarsland et al. 2007; Schrag et al.

2006). Moreover, this aspect generates consequent con-

cerns on whether and when to stop this therapy, especially

if the patient is demented, with severe disequilibrium or

already bedridden, as outlined by Nyholm (Nyholm et al.

2012) in a recent review.

Therefore, observing these consecutive 28 patients, it is

evident how our approach to this therapeutic option was

modified: at the beginning of our experience, we consid-

ered this opportunity for old and cognitive impaired

patients often with severe H&Y stage (4 or 5), whereas in

the last years we noted that the best long-term results were

obtained in younger patients with less severe stage of

disease and comorbidity. A similar reappraisal is now

taking place for DBS inclusion criteria (Schuepbach et al.

2013; Volkmann et al. 2013) and probably the time has

come to hypothesize this treatment earlier in disease course

also for Duodopa patients (Johansson and Nyholm 2012).

Hence, further studies evaluating the cost/effectiveness of

M. Sensi et al.

123

earlier introduction of duodenal levodopa infusion are

required (Nyholm 2012; Johansson and Nyholm 2012;

Nyholm et al. 2012; Antonini et al. 2013).

Conflict of interest Mariachiara Sensi has received lecture fees and

payment for advisory boards from Abbott, Boehringer Ingelheim,

Lundbeck, UCB, Merck Serono. Rocco Quatrale has received lecture

fee, payments for advisory boards from Abbott, Boehringer Ingel-

heim, Novartis, Lundbeck, UCB, Merck Serono, Chiesi, GSK. Fran-

cesca Preda has received consultancy fees from Medtronic.

References

Aarsland D, Andersen K, Larsen JP et al (2004) The rate of cognitive

decline in Parkinson disease. Arch Neurol 61:1906–1911

Aarsland D, Brønnick K, Ehrt U (2007) Neuropsychiatric symptoms

in patients with Parkinson’s disease and dementia: frequency,

profile and associated care giver stress. J Neurol Neurosurg

Psychiatry 78:36–42. doi:10.1136/jnnp.083113

Antonini A, Tolosa E (2009) Apomorphine and levodopa infusion

therapies for advanced Parkinson disease: selection criteria and

patient management. Expert Rev Neurother 9(6):859–867

Antonini A, Isaias IU, Canesi M et al (2007) Duodenal levodopa

infusion for advanced Parkinson’s disease: 12-month treatment

outcome. Mov Disord 22:1145–1149

Antonini A, Mancini F, Canesi M et al (2008) Duodenal levodopa

infusion improves quality of life in advanced Parkinson’s

disease. Neurodegenerative Dis 5(3–4):244–246

Antonini A, Odin P, Lopiano L (2013) Effect and safety of duodenal

levodopa infusion in advanced Parkinson’s disease: a retrospec-

tive multicenter outcome assessment in patient routine care.

J Neural Transm. doi:10.1007/s00702-013-1026-9

Chaudhuri KR, Jose Rojo JM, Schapira AHV et al. (2013). A proposal

for a comprehensive grading of Parkinson’s disease severity

combining motor and non-motor assessments: meeting an unmet

need. PLOS ONE www.plosone.org 8 (2): e57221

DANMODIS, Danish Movement Disorder Society; SWEMODIS,

Swedish Movement Disorder Society (2008) Duodopa: Scandi-

navian Consensus I. Treatment with levodopa/carbidopa gel

(Duodopa) in patients with Parkinson’s disease, Available at:

http://info.parkinsonforbundet.se/wpcontent/uploads/duodopa

konsensus-20081.pdf

Defer GL, Widner H, Rose- M et al (1999) Core assessment program

for surgical interventional therapies in Parkinson’s disease

(CAPSIT-PD). Mov Disord 14(4):572–584

Devos D (2009) Patient profile, indications, efficacy and safety of

duodenal levodopa infusion in advanced Parkinson’s disease.

Mov Disord 24(7):993–1000

Eggert K, Sckwrader C, Habner M et al (2008) Continuous jejunal

levodopa infusion in patients with advanced Parkinson disease:

practical aspects and outcome of motor and non-motor compli-

cations. Clin Neuropharmacol 31(3):151–166

Fasano A, Ricciardi L, Lena F et al (2012) Intrajejunal levodopa

Infusion in advanced Parkinson’s disease: long-term effects on

motor and non-motor symptoms and impact on patient’s and

caregiver’s quality of life. Eur Rev Med Pharmacol Sci

16(1):79–89

Fernandez HH, Vanagunas B, Odin P et al (2013) Levodopa–

carbidopa intestinal gel in advanced Parkinson’s disease open-

label study: Interim results. Parkinsonism Relat Disord

19:339–345

Glozman JM, Bicheva KG, Fedorova NV (1998) Scale of quality of

life of care-givers (SQLC). J Neurol 245(Suppl 1):S39–S41

Hindle V (2013) The practical management of cognitive impairment

and psychosis in the older Parkinson’s disease patient. J Neural

Transm 120:649–653

Honig H, Antonini A, Martinez PP et al (2009) Intrajejunal levodopa

infusion in Parkinson’s disease: a pilot multicenter study of

effects on non-motor symptoms and quality of life. Mov Disord

24(10):1468–1474

Johansson A, Nyholm D (2012) Continuous delivery of energy or L-

dopa: identifying advantages and limitations of DBS and

levodopa–carbidopa intestinal gel in the absence of head-to-

head comparisons. Basal Ganglia 2:221–226

Kulisevsky J, Pagonabarraga J, Pascual-Sedano B et al (2008)

Prevalence and correlates of neuropsychiatric symptoms in

Parkinson’s disease without dementia. Mov Disord

23(13):1889–1896

Kurth MC, Tetrud JW, Tanner CM et al (1993) Double-blind,

placebo-controlled, crossover study of duodenal infusion of

levodopa/carbidopa in Parkinson’s disease patients with ‘‘on–

off’’ fluctuations. Neurology 43:1698–1703

Nyholm D (2012) Duodopa treatment for advanced Parkinson’s

disease: a review of efficacy and safety. Parkinsonism Relat

Disord ep 18(8):916–929

Nyholm D, Askmark H, Gomes-Trolin C et al (2003) Optimizing

levodopa pharmacokinetics: intestinal infusion versus oral

sustained-release tablets. Clin Neuropharmacol 26(3):

156–163

Nyholm D, Nilsson Remahl AIM, Dizdar N et al (2005) Duodenal

levodopa infusion monotherapy vs oral polypharmacy in

advanced Parkinson disease. Neurology 64(2):216–223

Nyholm D, Lewander T, Johansson A et al (2008) Enteral levodopa/

carbidopa infusion in advanced Parkinson disease: long-term

exposure. Clin Neuropharmacol 31(2):63–73

Nyholm D, Klangemo K, Johansson A (2012) Levodopa/carbidopa

intestinal gel infusion long-term therapy in advanced Parkinson’s

disease. Eur J Neurol. doi:10.1111/j.1468-1331.201203679.x

Palhagen SE, Dizdar N, Hauge T et al (2012) Interim analysis of long-

term intra-duodenal levodopa infusion in advanced Parkinson

disease. Neurol Scand. doi:10.1111/j.1600-0404.2012.01689

Puente V, De Fabregues O, Oliveras C et al (2010) Eighteen month

study of continuous intra-duodenal levodopa infusion in patients

with advanced Parkinson’s disease: impact on control of

fluctuations and quality of life. Parkinsonism Relat Disord

16(3):218–221

Reddy P, Martinez-Martin P, Rizos A, Martin A et al (2012)

Intrajejunal levodopa versus conventional therapy in Parkinson

disease: motor and non-motor effects. Clin Neuropharmacol

35(5):205–207

Robottom BJ, Gruber- Baldini AL, Anderson KE et al (2012) What

determines resilience in patients with Parkinson’s disease?

Parkinsonism Relat Disord 18:174–177

Sanchez-Castaneda C, Jaume Campdelacreu J, Miro J et al (2010)

Cognitive improvement after duodenal levodopa infusion in

cognitively impaired Parkinson’s disease patients. Prog Neuro

Psychopharmacol Biol Psychiatry 34:250–251

Santos-Garcia D, Anon MJ, Fuster-Sanjurjo L, de la Fuente-

Ferna0ndez R (2012) Duodenal levodopa/carbidopa infusion

therapy in patients with advanced Parkinson’s disease leads to

improvement in caregivers’ stress and burden. Eur J Neurol

19(9):1261–1265

Schrag A, Hovris A, Morley D et al (2006) Caregiver-burden in

Parkinson’s disease is closely associated with psychiatric

symptoms, falls and disability. Parkinsonism Relat Disord

12:35–41

Schuepbach WMM, Rau J, Knudsen K et al (2013) Neuro-stimulation

for Parkinson’s disease with early motor complications. New

Engl J Med 368:610–622

Emerging issues in Duodopa patients’ selection criteria

123

Valldeoriola F, Puig-Junoy J, Puig-Peiro R, Workgroup of the SCOPE

study (2013) Cost analysis of the treatments for patients with

advanced Parkinson’s disease: SCOPE study. J Med Econ

16(2):191–201. doi:10.3111/13696998.2012.737392 (Epub 2012

Oct 29)

Volkmann J, Albanese A, Antonini A et al (2013) Selecting deep

brain stimulation or infusion therapies in advanced Parkinson’s

disease: an evidence-based review. J Neurol. doi:10.1007/

s00415-012-6798-6

Westin J, Nyholm D, Groth D et al (2006) Outcome prediction of

enteral levodopa/carbidopa infusion in advanced Parkinson’s

disease. Parkinsonism Relat Disord 12:509–513

Zibetti M, Merola Z, Ricchi V et al (2013) Long-term duodenal

levodopa infusion in Parkinson’s disease: a 3-year motor and

cognitive follow-up study. J Neurol 260(1):105–114

M. Sensi et al.

123